The applicant for this K02 award is a thrice-refunded NIDA grantee in nosology. The award will equip him with enhanced nosologic research skills, promoting his ability to contribute to the better understanding of the etiology, epidemiology, treatment, and long-term course of substance use disorders. This revision responds in detail to comments from an earlier round of reviews and features: (a) inclusion of three new Consultant/Collaborators with skills in non-alcohol drug nosology and epidemiology, (b) inclusion of an epidemiologic dataset, the National Comorbidity Study, in the secondary analysis plans, (c) inclusion in the Career Development Plan of training in the analysis of epidemiologic data, (d) integration of non-alcohol drug literatures in the Specific Aims and Hypotheses and Data Analyses, (e) discussion of limitations of the statistical techniques used, and others. Supported activities will be conducted with the assistance of a cadre of Consultant/Collaborators expert in the skills areas identified, and the award will secure for the candidate protected time to reposition for the next decade an ongoing and productive line of research. The first proposed K02 career development activity consists of acquiring advanced quantitative skills carefully chosen for both innovativeness and applicability to stubborn nosologic problems at the subcriterion, criterion and composite algorithm levels. These include (f) IRT-based methods, (g) event-history analysis, (h) ROC and QROC analysis, (i) taxometric methods (MAXCOV/HITMAX and catastrophe modeling),and (j) epidemiologic assessment strategies. The Research Plan will apply these methods to secondary analyses of four existing NIH datasets. These analyses will bring existing DSM and ICD diagnostic algorithms for substance use disorders to higher performance levels through three strategies - criterion weighting, criterion redistribution, and resetting diagnostic thresholds with empirical techniques. This work will lead in a following phase to development of diagnostic models alternative to DSM/ICD such as drug-customized algorithms and the "Withdrawal-Gate Model," with better performance parameters and firmer links to biological and human developmental factors. A second career development activity consists of acquiring advanced field research skills in community sampling and in perfecting research interviews and queries tailored to the improved diagnostic algorithms designed in the earlier research phases. This is necessary for the preparation, toward the close of the K02 award, of an advanced nosologic field study, not covered here, that will feed results immediately in to the DSM-V and ICD-11 development processes that will then be under way.